Since it is apparent that there is a limit in scaling-down a non-volatile semiconductor memory in related arts in which silicon (Si) is used for a front end process, examination has been started on various kinds of non-volatile semiconductor memories in which new materials are used, such as a resistive random access memory (ReRAM), a phase change random access memory (PCRAM), and a magnetic random access memory (MRAM).
Among these non-volatile semiconductor memories, for example, a ferroelectric random access memory (FeRAM) in which a ferroelectric substance is used has not been put into a practical use because of various factors, except that the FeRAM is used for a small-sized memory such as a radio frequency identification (RFID) card for the purpose of power saving. Examples of the factors are an inclusion of material difficult to handle such as lead, and difficulty in film-thinning due to a size effect, and so on.
Particularly, since ferroelectric substances having low affinity with a semiconductor process such as lead, bismuth, and barium are needed to be used, there is the scaling-down limit while the process is becoming more complex. Further, an application range is also limited because manufacturing cost can be hardly reduced.